1. Field of the Invention
The present invention is generally related to a method for measuring a net charge density of a membrane and an apparatus thereof, and more particularly to a method for measuring a net charge density of a porous membrane such as a microfiltration, ultrafiltration or nanofiltration membrane and an apparatus thereof.
2. Description of the Prior Art
The characteristic parameters of a membrane generally can be categorized into performance parameters, morphology parameter, and charge parameters. In the charge parameters, the zeta potential of a membrane usually is measured by the streaming potential method or electroosmosis method and its influence to the filtration performance of the membrane is analyzed but the net charge density inside the membrane is usually uncertain.
Particularly, regarding nanofiltration (NF) membranes, the key of the species rejection mechanism of a nanofiltration (NF) membrane is the electrostatic repulsion between species and the NF membrane. The electrostatic repulsion force is an important factor in controlling the filtration performance. The charge parameters used to evaluate the filtration performance of the NF membrane and the ion exchange membrane are usually estimated from the net charge density per unit volume or area of a membrane. According to the prior art, the value of the net charge density of a NF membrane per unit volume is usually acquired by the estimation from the experimental result accompanying with the theoretical simulation. Since most of the NF membranes are composite membranes, the pore diameter is very small and thus overlapping of electric double layers is very serious to thereby cause the difficulty in measuring the real zeta potential of the surfaces of the pore walls of the NF membrane.
Regarding the estimation of the net charge density per unit volume, Ta-Shung et al. (Tai-Shung, C., J. Lv and W. Kai Yu, “Investigation of amphoteric polybenzimidazole (PBI) nanofiltration hollow fiber membrane for both cation and anions removal”, J. Membrane Sci., 310, 557-566 (2008)) use Speigler-Kedem model (SKM) together with Teorell-Meyer-Sievers model (TMS) to estimate the net charge density per unit volume for PBI nanofiltration (NF) membrane. SKM is used to describe the relation between solution flux and solute flux while TMS is used to describe the relation of ion permeation. Such a method accompanying with the result of the experimental blocking rate calculates the reflection coefficient of the membrane to thereby estimate the charge quantity of the membrane per unit volume. The result indicates that the effective net charge density per unit volume of the membrane increases with the increase of the ionic concentration.
On the other hand, Bandini et al. (Bandini, S., D. Jennifer and V. Daniele, “The role of pH and concentration on the ion rejection in polyamide nanofiltration membranes”, J. Membrane Sci., 264, 65-74 (2005)) use DSPM-DE (Donnan Steric Pore Model (DSPM) and Dielectric Exclusion) to estimate the net charge density per unit volume for the DK02 nanofiltration (NF) membrane (OSMONICS). The experiment of filtration of a sodium chloride aqueous solution is performed and the blocking rate obtained from the experiment accompanying with the theoretical model and the charge neutrality balance condition is used to estimate the net charge density per unit volume. The result shows that the charge density increases with the increase of the concentration of salts. The same result is obtained even for different pH values. Then, Bandini et al. (Bandini, S. and C. Mazzoni, “On nanofiltration Desal-5 DK performances with calcium chloride-water solutions”, Separation and Purification Technology, 52, 232-240 (2006)) also use DK02 nanofiltration (NF) membranes in the CaCl2 circulating sweeping filtration experiment and then the volumetric net charge density of the NF membrane is estimated from the ion blocking experimental result. The estimation result indicates that the electrical potential of the membrane is increased by ion adsorption and due to the competition between the adsorption of calcium ions and the adsorption of chloride ions when the concentration is increased to a threshold level the isoelectric point of volumetric charges of the membrane appears at two different concentrations.
Therefore, when, according to the prior art, the method of utilizing the experimental result of the ion blocking rate together with the theoretical model is applied to simulate the net charge density, the charge density is increased with the increase of the concentration of ions. In addition, it is found that larger deviation in estimation will appear if different methods are used based on various ionic radii defined. Thus, it is important to effectively measure a net charge density per unit volume and a more reliable measurement method is required.